INDEX

Explanations

mentions of specific historical eras

oai_token-act-pair · gpt-3.5-turbo

references to different historical periods or eras

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

amina

-0.84

acts

-0.81

lations

-0.75

ramid

-0.74

Ĥ¬

-0.74

portion

-0.72

Eva

-0.72

gments

-0.70

ergus

-0.70

olulu

-0.69

POSITIVE LOGITS

era

0.98

Era

0.94

UTC

0.84

 epoch

0.83

icals

0.82

 nostalg

0.81

 msec

0.72

Age

0.70

 Frieza

0.69

osaurs

0.68

Activations Density 0.017%

mentions of specific historical eras

references to different historical periods or eras

No Comments

No Known Activations